Object recognition and path-drawing for a mobile robot

Abstract

In this paper we propose a vision algorithm to object recognition during a mobile robot's journey in a controlled environment. Furthermore, another algorithm was used to drawing the path of the robot, which is based on the evaluation of a vision algorithm to provide results. The first algorithm uses color processing techniques to achieve an image thresholding and a Harris corner detector to recognize objects in a scene. A graphical user interface has been developed to perform the image analysis and the path-drawing in real time. These algorithms, working together, provided 97% as recognition rate.